The present invention relates to the dithering of motion sensors such as ring laser gyroscopes and mechanical gyroscopes.
A ring laser gyroscope is a laser apparatus having a ring type resonant cavity which may be more simply referred to as a ring resonator. The ring resonator is commonly constructed of a block of glass or glass ceramic having a plurality of interconnecting passages in the shape of a closed loop path such as, for example, a triangular or rectangular path. Laser beams are directed around the path by suitable mirrors appropriately positioned at the intersections of pairs of the interconnecting passages.
In ring laser gyroscopes, there are commonly two laser beams traveling in opposite directions (clockwise and counterclockwise) relative to each other around the closed loop path formed by the ring cavity. The positioning of the mirrors at the corners of the closed loop path direct the laser beams through the passages of the resonant cavity. A mirror at one of the corners is partially transmissive so that a portion of each of the counter-propagating beams is passed to a readout assembly. Some examples of ring laser gyroscopes are shown and described in U.S. Pat. No. 3,373,650 and U.S. Pat. No. 3,467,472 issued to Killpatrick, and in U.S. Pat. No. 3,390,606 issued to Podgorski.
A source of error in the output of a ring laser gyroscope is xe2x80x9clock-in.xe2x80x9d At rotation input rates below some critical value called the lock-in threshold or the lock-in rate, the counter-propagating beam frequencies synchronize to a common value resulting in a zero frequency difference between the counter-propagating beams. Because the frequency difference between the beams is used to determine the rotation rate of the ring laser gyroscope, a zero frequency difference at low rotation rates due to lock-in erroneously indicates no rotation.
To maintain a frequency difference between the counter-propagating beams at low rotation rates and thereby avoid lock-in, ring laser gyroscopes have been biased into oscillation about their input axis. Such biasing is shown and described in the aforementioned U.S. Pat. No. 3,373,650. This bias oscillation of a ring laser gyroscope is referred to as dither and is commonly provided by a dither motor which rotates the gyroscope block relative to an inertial platform, as further shown and described in the aforementioned patent. The oscillating rotation bias results in rotation rates that are higher than the lock-in rate for a majority of the operating time.
Typically, a dither motor is comprised of at least one piezoelectric transducer (PZT) attached to a corresponding one of the spokes of a dither spring as shown and described in U.S. Pat. No. 4,370,583 issued to Ljung. The dither spring is generally composed of a central member or hub which is in turn attached to an inertial platform. The spokes of the dither spring are attached at one end to the hub. These spokes extend radially from the hub and are attached at opposite ends to a toroidal rim which engages the gyroscope block.
A sinusoidal drive signal is applied to the aforementioned PZT. The PZT causes flexing of the spoke to which the PZT is attached. This flexing oscillates the rim relative to the hub and thereby rotationally oscillates the gyroscope block relative to the inertial platform. Additionally, as taught in the aforementioned U.S. Pat. No. 3,467,472, noise may be introduced to the sinusoidal signal to further decrease lock-in effects.
Usually, it is desirable to oscillate the gyroscope block, relative to the inertial platform, at the natural resonant frequency of the dither motor. To achieve oscillation at the resonant frequency, a dither sensor is commonly provided and typically comprises at least one PZT dither sensor which is attached to a spoke to sense motion of the gyroscope block relative to inertial platform motion. The output of the dither sensor is used to change the sinusoidal drive signal supplied to the dither motor such that oscillation at the resonant frequency results. Specifically, when the spoke flexes in the aforementioned manner, the PZT dither sensor deforms and produces a responsive output signal, thereby sensing flexing of the spoke. This output signal, or xe2x80x9cpick-offxe2x80x9d signal, is provided as an input to a feedback circuit. The feedback circuit controls the amplitude and/or frequency of the sinusoidal drive signal supplied to the dither motor such that the dither motor oscillates at or near its natural resonant frequency.
Unfortunately, dithering causes an error angle component in the gyroscope output (i.e., in the frequency difference between the counter-propagating beams), as noted by Killpatrick in Laser Gyro Dither Random Noise Proceedings of S.P.I.E., Meeting on Physics of Optical Ring Gyros, vol. 487 at 85-93 (1984). This error angle component, or noise, in the gyroscope output results in a rotation error in the output of the gyroscope. For a high frequency dithering rate, this error is mathematically represented in the aforementioned reference by the following equation:                               Δψ          ⁡                      (            t            )                          =                                            Ω              L                        ⁡                          (                              Kt                                  2                  ⁢                                      πΩ                    D                                                              )                                            1            /            2                                              (        1        )            
where xcex94"psgr"(t) is the error angle component in arc seconds, K is a gyroscope scale factor in arc seconds/cycle, xcexa9L is the lock-in rate in arc seconds/second, xcexa9D is a dither angular rotation rate in arc seconds/second, and t is operating time in seconds.
Dividing through by t in equation (1) to obtain the error angle component rate, also known as the random drift error, produces the following equation:                                           Δψ            ⁡                          (              t              )                                t                =                                            Ω              L                        ⁡                          (                              K                                  2                  ⁢                                      πΩ                    D                                    ⁢                  t                                            )                                            1            /            2                                              (        2        )            
By inspection of equation (2), it is evident that, when K, xcexa9L, and xcexa9D are held constant, the random drift error xcex94"psgr"(t)/t, or the total error rate in the gyroscope output, decreases with increasing operating time t. Consequently, random drift rate error decreases with increasing gyroscope operating time.
Alternatively, it is also evident in equation (2) that, during a fixed operating time t, an increase in dither angular rotation rate xcexa9D, with K and xcexa9L held constant, also decreases random drift error.
Calibration, or xe2x80x9calignment,xe2x80x9d of an inertial system using a ring laser gyroscope is usually performed by using the ring laser gyroscope to sense the rotation of the earth when the platform of the ring laser gyroscope is stationary except for the earth""s rotation. Rotation rate information derived from this sensing xe2x80x9calignsxe2x80x9d the inertial system because this information determines North (as well as South, East, and West). The accuracy of this alignment process depends upon the gyroscope""s ability to read the input rotational rate to an acceptable accuracy. For example, at a latitude of 45 degrees, the horizontal component of the earth""s rotation is approximately 10 degrees/hour in the North and South directions, and 0 degrees/hour in the East and West directions. A gyroscope error of 0.01 degrees/hour then produces an error in alignment of 1 milliradian (0.01/10) radians. The elapsed time necessary to align the gyroscope, known as the alignment time, is determined by the time it takes for the gyroscope error to reduce to a value, for example, of 0.01 degrees/hour.
Because the gyroscope is operating during alignment, t may also represent alignment time in equation (2). It is desired that the alignment time be very short. However, if t representing alignment time in equation (2) is made small, then a large random drift error over that time results.
Moreover, many navigation systems rely on global positioning signals derived from global positioning satellites to increase the accuracy of the position information derived from the inertial sensing provided by a ring laser gyroscope. Thus, the accuracy of a vehicle""s position can be increased by these global positioning signals. However, when such global positioning signals are not available, position information provided by a navigation system can contain a higher degree of error.
The present invention is directed, at least in one of its aspects, to an arrangement for reducing errors during periods of alignment and/or during periods when global positioning signals are not present.
Other types of motion sensors such as mechanical gyroscopes are also known. A mechanical gyroscope typically includes a mass that is suspended on a platform whose rotational rate is to be determined. The mass is dithered with a linear oscillatory motion. When a rotational rate is also imposed on the mass, a Coriolis force is generated that is perpendicular to both the linear oscillatory motion and the rotational rate.
This Coriolis force F is given by the following equation:
F=2mxcexa9xVxe2x80x83xe2x80x83(3) 
where m is the mass, xcexa9 is the rotational rate to be measured, V is the instantaneous velocity of the linear oscillatory motion described above, and x represents the mathematical cross product. As discussed above, the Coriolis force F causes the mass to move in a direction that is perpendicular to both the rotational rate and the instantaneous velocity of the linear oscillatory motion. The amplitude of this movement is sensed to provide an output proportional to the rotational rate.
As in the case of ring laser gyroscopes, mechanical gyroscopes are also aligned. The present invention is directed, at least in another one of its aspects, to an arrangement for increasing sensitivity of a mechanical gyroscope by increasing dither amplitude so that the mechanical gyroscope can sense smaller rotation rates during periods of alignment.
In accordance with one aspect of the present invention, a method of dithering a motion sensor comprises the following: dithering the motion sensor with a signal having a first rate amplitude when alignment is not being performed; and, dithering the motion sensor with a signal having a second rate amplitude when alignment is being performed, wherein the second rate amplitude is greater than the first rate amplitude.
In accordance with another aspect of the present invention, a method of dithering a motion sensor comprises the following: dithering the motion sensor with a signal having a first rate amplitude when a global positioning signal is available; and, dithering the motion sensor with a signal having a second rate when the global positioning signal is not available, wherein the second rate amplitude is greater than the first rate amplitude.
In accordance with yet another aspect of the present invention, a method of dithering a motion sensor comprises then following: dithering the motion sensor with a signal having a first rate amplitude if a global positioning signal is available; dithering the motion sensor with a signal having a second rate amplitude when alignment is not being performed; dithering the motion sensor with a signal having a third rate amplitude if the global positioning signal is not available; and, dithering the motion signal with a signal having a fourth rate amplitude when alignment is being performed, wherein the third rate amplitude is greater than the first rate amplitude, and wherein the fourth rate amplitude is greater than the second rate amplitude.
In accordance with still another aspect of the present invention, a dither system for dithering a motion sensor comprises a dither sensor, a dither motor, and a controller. The dither sensor senses dithering of the motion sensor. The dither motor dithers the motion sensor. The controller has first and second inputs and an output. The output is coupled to the dither motor, and the first input is coupled to the dither sensor. The controller drives the dither motor with a signal having a first dither rate amplitude in response to the first input, and the controller drives the dither motor with a signal having a second dither rate amplitude in response to the second input. The second dither rate amplitude is higher than the first dither rate amplitude.
In accordance a further aspect of the present invention, a method of dithering a motion sensor comprises the following: applying an amplitude control signal to a dither system coupled to the motion sensor such that the motion sensor oscillates in a first mode during a selected time, wherein the amplitude control signal has a first amplitude during the first selected time; adjusting the amplitude control signal after the selected time such that the first amplitude is decreased to a second amplitude; and, applying the amplitude control signal having the second amplitude to the dither system coupled to the motion sensor such that the motion sensor oscillates in a second mode.